Connect with us

Help, any gurus with alternator experience or knowledge?

Discussion in 'Electronic Repair' started by default, Jul 8, 2006.

Scroll to continue with content
  1. default

    default Guest

    The quandary

    I'm rewinding an automobile style alternator rotor. I want to use
    some "Tefzel" insulated wire I have on hand, instead of magnet wire, I
    have to special order. I wound a test coil and it fits and looks like
    it will work.

    My concern is that I have no idea how much current it takes to excite
    the field and if the potting compound will survive the heat. The
    Tefzel coil has the same DCR (5 ohms) as the original magnet wire, but
    it is two gauges thinner (went from 22 AWG to 24 AWG). So I would
    assume that it will dissipate more power to achieve the same ampere
    turns in the field.

    I can't test it very well without potting the coil and epoxying it
    into the pole pieces, mounting it to the engine, etc. - and if it
    turns out to be bad, it is a real bitch to pull apart and do over with
    different wire.

    I figure the excitation power probably drops with speed of rotation -
    alternator voltage output tracks speed so it should need less
    excitation as speed increases - and the frequency goes up so inductive
    reactance also increases(?)

    So, I'm thinking worst case is probably close to idle speed. To
    further complicate that idea, excitation also has to track speed to
    some extent, since it is derived from an extra set of diodes from the
    rectifier - lower speed means less current/voltage to work with.

    I tried powering the coil with a dc supply and pushed 2.5 amps through
    it for 3 hours - no idea how hot the coil was, but the area between
    the coil and pole pieces was 70 degrees F over ambient - around 150 F.
    On the engine, it is driven directly off the crankshaft and probably
    has an ambient of closer to 170 F - enclosed with no ventilation just
    conduction and radiation cooling, and whatever air the rotor itself
    stirs up.

    I can do some empirical testing with a sacrificial coil when the vinyl
    ester resin gets here.

    Anyone with experience/ideas in rewinding rotors and do you think this
    should work? Smaller gauge wire - same DCR, but lower ampere turns
    and consequently more power used to cause more heat and excite the

    The potting resin is supposed to be good for ~240 F so I might be
    pushing the limit there.

    The whole story:

    Five years ago my alternator failed. The rotor had shorted -
    resistance of the coil would vary from point five to five ohms - five
    ohms is supposed to be typical. Replacement rotor $350 . . . with no
    guarantee that it wouldn't fail like the original in two years . . .

    The regulator is the common type usually used with excited field
    alternators - a two transistor circuit that pumps voltage to the coil
    when the battery drops below the set point. The voltage that goes to
    the rotor (rotating field) is derived from an extra three diodes on
    the positive of the six diode, three phase rectifier - so it is
    isolated from the battery.

    One effect of using extra diodes is that when the output of the
    alternator drops (when the field is shorted, for instance) the
    excitation current is also lower - doesn't do much to charge the
    battery, but it doesn't kill the battery in an effort to excite a
    shorted field, and doesn't kill the regulator pass transistor. A good
    design . . .

    The original coil failed because the enamel on the magnet wire and or
    varnish holding it together failed (probably because of heat or
    vibration - at least that's what the wire looked like). It was a self
    supporting coil - made in a mold and had no bobbin.

    I didn't have the stuff to make a self supporting coil so I made a
    bobbin out of very thin two sided epoxy pcb material, and insulated
    the inside with pieces of thin Mylar plastic. Wound a layer - painted
    it with epoxy and built up the entire coil that way. It lasted 5
    years and then failed because the lead wires to the coil opened - The
    wires were in a sleeve of Teflon spaghetti and probably opened due to
    metal fatigue - that's what the ends looked like - when you bend a
    wire back and forth until it breaks. I repaired one open to the
    finish end and it worked for a few weeks and then the start end also
    opened. The coil is pretty much a goner now - the clear epoxy shows
    the wire to be in excellent shape - no charring like the OEM part.

    I want to wind another coil but would like to avoid using a bobbin
    since that took me over a day to construct with hand tools, and the
    bobbin didn't survive pulling apart the pole pieces.

    So I found some wire wrap wire with "Tefzel" insulation and wound a
    coil with that on the mandrel that supported my original bobbin. I
    secured it into a toroid shape using nylon lacing cord. Fits the pole
    pieces and looks like it will work. Plan B was to serve leads to it
    made of fine braid - to avoid metal fatigue and sleeve it in cambric
    spaghetti then dip it in vinyl ester resin, epoxy that assembly into
    the pole pieces and reassemble the rotor.

    Plan A is to laboriously construct a new bobbin (1+ day of effort) and
    order the right gauge wire and do the wind - epoxy routine, then serve
    leads made of braid instead of wiring directly to the slip rings. A
    lot of work.
  2. Rich Grise

    Rich Grise Guest

    Could you temporarily stick it in place with hot glue, just for testing
    purposes? Hot glue should just be able to just peel right off between
    tests and so on.

    Good Luck!
  3. default

    default Guest

    Probably long enough to start the engine and measure the current. I
    was sort of hoping to get it to work without a test except perhaps a
    coil alone - with a fixed dissipation.

    Even the hot melt adhesive isn't a pleasant idea - run long enough to
    get real data and I'd probably have the adhesive slinging off into the
    stator or space between the rotor and stator - too risky to chance
    that - adhesive being ground up would take out the brushes - unlike
    auto alternators, this one has axial slip rings - any bit of epoxy etc
    that comes off eats the brushes.
  4. jasen

    jasen Guest

    I'm no guru but:

    if the wire is thinner, but the coil the same dimension then you
    theoretically have more turns - this means for the same current you have
    more magnetic field.

    if the coil is the same resistance as the old one then you will have the
    same max current, and the same worst-case resistive heating as the old coil.

    if the conductor is thinner but the insulated wire is thicker you have fewer
    turns and threfore a less-effective altenator.
    the coil is fed DC, inductance doesn't enter into it.
    those diodes don't give a greatly elevated voltage, they're mainly to
    provide a way to power the generator warning light.
    your altenator doesn't vent slots or a fan behind the pulley?
    maybe you can epoxy in screw terminals this time incase the leads fail.

    maybe you could use old CD-Roms (the shiny can be removed using a metal pot
    scourer) for the ends of the bobbin and a dowel (or threaded rod?) wrapped
    in paper and cling-wrap for the centre?

    hmm, if I could fix a 3-jaw chuck to the back of a sewing machine that'd
    make a good tool start for winding magnet coils
  5. Vaughn

    Vaughn Guest

    Again no guru, but from what I remember of winding my own transformers,
    more coil turns equals greater voltage, so this alternator MAY produce a
    greater voltage for a given rotational speed.
    But as the Alternator produces an Alternating current, don't they also
    work as a Bridge Rectifier?

    Like I said NO GURU, but its my pennys worth ;-)

    Best of luck with the build


  6. Automotive alternators are usually three phase with six diodes. this
    helps reduce the ripple current in the charging system, and reduces
    filtering requirements for the electronics.

    Service to my country? Been there, Done that, and I've got my DD214 to
    prove it.
    Member of DAV #85.

    Michael A. Terrell
    Central Florida
  7. Jim Land

    Jim Land Guest

    Anytime you rewind a motor or alternator, you want to use the same gauge
    wire. Your alternator was designed with 22 AWG, for the number of
    turns, the dc resistance, the current through it, the power dissipation,
    and the space available. If it would work just as well with 24 AWG,
    they would have used 24 AWG originally. If you're going to all the
    trouble of winding these coils, placing them, and wiring them up, you
    might as well use the right size wire to start with!
  8. Ban

    Ban Guest

    When your wire is thinner but has the same resistance, it will be
    shorter(40%) and hence the field will be weaker, probably less than half. It
    also means it will heat up much more for 2 reasons: The diameter is smaller
    and the insulation is much thicker, so the heat cannot be transferred to the
    metal or air so easily, builds up in the center and can blow up the sleeve
    of epoxy. Another remark: how well do you think your insulation is sticking
    to *any* glue? An additional property of PTFE is its tendency to "flow"
    aggravated by vibration, temperature and pressure, all of which are present
    in your application.
    Back EMF increases linearly with speed, but since you have less windings,
    the current will be higher.
    forget about the tefzel for this app and get the right magnet wire suitable
    for high temp.
  9. me

    me Guest

    Quit bloody wating your time(and likely effort) and get one at a salvage
  10. default

    default Guest

    Thanks, you've given me some ideas.

    I kinda lied about the resistance - my repair to the original
    measures 4.5 ohms and the Tefzel coil measures 3.8 . Turns out my DVM
    isn't great in the low ohms range - two different days and two
    different readings.

    The Tefzel wire does have what appears to be a silver plating if that
    is enough to enter into it. What I can do is measure the thickness of
    the ETFE and compare that to an enamel wire table and see how many
    turns should fit in the same cross sectional area and pin down the
    turns difference with some accuracy. My repair winding was relatively
    sloppy and wasn't "perfect lay" by any means.

    PTFE FEP PFA are in the same class when it comes to physical
    properties ETFE is Tefzel and is in a different class - twice the
    tensile strength, 20% harder, better compressive strength, similar
    flexural strength. Lower temperature rating for the physical
    properties than FEP (260 C) types but good to 155 degrees C.

    Back EMF in the excited field rotor? I didn't know there was any.

    I'm not convinced current will be higher - the excitation current is
    directly derived from the alternator output - it isn't fixed - it
    doesn't come from the battery. A shorted or nearly shorted rotor is
    less effective for output at the same rotational speed so it has less
    voltage to send to the rotor - that's the reason for an extra three
    diodes on the regulator - not for the warning light as one poster
    suggested (it has no warning light). There is a fixed dropping
    resistor that provides a tickle of current from the battery to get
    things started.
    Price of magnet wire in small quantities has quadrupled in the last
    year and doubled in large quantities - place I bought one pound from
    for $12 now wants $35 for a half pound (8 ounces) and I need about 12
    ounces. At that rate it would probably be cheaper to buy a 6 pound
    reel of the stuff.

    Yeah - you're right to point out ETFE won't stick to epoxy, to work it
    has to penetrate between the wires - I'm using vinyl ester resin,
    similar to polyester resin (water thin compared to epoxy and good
    wetting properties) higher temperature rating than epoxy. I'll use
    epoxy to hold the coil in place or thickened vinyl ester resin for

    Blow it apart due to thermal expansion? That should be no problem to
    discover with empirical testing - vinyl ester resin sets up hard -
    but it isn't glass hard - just hard compared to unfilled epoxy. The
    coil will always have some air trapped in the center of the windings
    unless I can figure out how to turn my mandrel into a mold (I'm
    working on that now)

    Buying wire is always an option - but I'll kick this around some more
    before I give up. I have two weeks to get it working.

    Thanks for the reply.
  11. default

    default Guest

    Would that it was that simple. These things are scarcer than hen's
    teeth, and all have a reputation for failing as originally made.

    There is some emotional satisfaction to be derived from solving
  12. default

    default Guest

    You are forgetting the extra diodes in the regulator - they provide
    the current to the field rotor - not the battery. The warning light
    on automobiles serves two functions - it provides the "tickle" of a
    few hundred milliamps get things started as well as indicate that the
    alternator is working. Some autos won't charge the battery at all
    when the warning light burns out. My system has no warning light just
    a small resistor to start things off.
    Fed pulsating DC albeit three phase rectified DC, inductance may be a
    factor. I don't know. Right now it is an unknown variable - when I
    switch on a load the voltage drops for ~100+ milliseconds and
    overshoots again when I switch it off - if the regulation were better
    I'd say inductance didn't matter.
    No fan, no pulley, driven - bolted directly to the crankshaft. No
    venting what so ever - no way for rain to get in. Conduction and
    radiation and convection around the cast aluminum cover.
    That's a good idea. screw terminals of brass tabs imbedded in the
    epoxy to solder to
    Here if I send a mini CD through the mail in a regular envelope - it
    comes back with the shiny side in the bottom of the envelope - Have to
    see how the post office does that - no scouring necessary, perfectly
    Actually I have a crude coil winder. I have a large tape recorder,
    reel drive, DC, permanent magnet, motor. It has a taper on the end
    of its 2" long shaft. To the end of the motor I can bang on a block
    of wood and turn it down with a chisel and rasp to any diameter from
    1/2" to 5" I already built a three part mandrel to wind the coil out
    of wood - the tricky part is figuring out how to get some plastic on
    there that will act as a mold release and mold a self supporting coil.

    The coil winder was made for winding long Tesla coils and I did use it
    for a large induction coil. Works very well with an adjustable DC
    supply to power it. Crank up enough voltage to pull wire from a
    supply reel and guide it on by hand. Lower the voltage to keep
    tension on it to paint with varnish, epoxy, or to get a beer. Paint a
    Tesla coil with epoxy and set it to rotate slowly while it cures and
    it is possible to get a really smooth finish.

    The winding lathe has a tail stock made from a carriage bolt and is
    threaded through a long nut mounted in blocks of wood - for Tesla
    coils - short coils go on the mandrels mounted to the motor shaft and
    turned to diameter.

    I have some of those mini CDR discs and they are the ideal diameter
    for side plates (have to bore out the center to about 1-5/8") I don't
    know that I'd trust the poly carbonate plastic at those temperatures
    for a permanent bobbin - a tin can lid would make better sense there.

    I'm toying with the idea of using a couple of polyethylene recyclable
    lids like one sees on re closable food cans to make temporary side
    plates. Coat them will silicone wax and epoxy the coil together and
    then remove it from the mold. The spindle (hub/center) could be
    wrapped with Teflon pipe joint tape and making the side plates
    undersize and sliding it over the tape may give a liquid tight seal to
    make a self-supporting coil. Then that coil could be further
    insulated with tape or dipping in resin.

    Thanks for the reply - more ideas to think about
  13. default

    default Guest

    I wind transformers too. If one winds the stator (non movable outside
    part on my alternator) with more turns, the voltage will increase for
    the same rotational speed.

    We are talking about a DC electromagnet here - the rotating field --
    magnetizing a hunk of cast iron (pole pieces). I ran some calculations
    on wire size and cross sectional area of the coil. All things being
    equal - same magnet wire insulation - the resistance goes up (as wire
    gets thinner) the current goes down (greater resistance) but the
    magnetic field strength stays the same if the cross section (filled
    with wire) stays the same.

    I'm contemplating changing the size of the insulation with a (possibly
    silver plated) copper wire and increasing insulation thickness and
    decreasing wire diameter two gauges - that's what gives me pause.
    Three phase delta connected output that goes to six diodes for primary
    rectification and three additional (smaller) diodes to provide field
    excitation independent of the battery.
    Thanks - I enjoy tinkering with it - but don't want to get stuck
    hundreds of miles from home with no way to repair it.
  14. default

    default Guest

    Oh, I don't know . . . where's your sense of adventure man?

    All engineering is a matter of compromise. You assume the factory got
    it right

    - the original lasted 2 years my first repair lasted 5. It would
    still be working if I'd thought to serve some stranded wire to the
    start and finish ends and not used Teflon tubing to support the wire.

    There are some alternators that last until the brushes die from wear
    or the diodes rust through - this alternator has a very high, well
    documented, rotor failure rate. I'm assuming the factory did not get
    it right.

    I ran some calculations on wire size and cross sectional area of the
    coil. All things being equal - same magnet wire insulation - the
    resistance goes up (as wire gets thinner) the current goes down
    (greater resistance) but the magnetic field strength stays the same if
    the cross section (filled with wire) stays the same.

    The power dissipation goes down as the wire gets thinner, lower
    current producing the same ampere turns in the same available space
    assuming the voltage stays constant - higher efficiency.

    I'm committing two sins here - going down in wire diameter and
    increasing insulation thickness - fewer turns of thinner wire.
    Conductivity may be better because it looks like silver plating on the
    wire. I wish I'd measured the 1,000 foot reel before I started using
    it . . . but I bought it for Tesla coils not alternator rotors.

    The downside of using very thin wire is it is more likely to break
    under stress, physical or thermal cycling. At some point wire
    insulation may take up more room than wire itself with very thin wire
    gauges - but within the practical range of several gauges it shouldn't
    matter all that much. Theoretically.

    I would expect the output to go up slightly with thinner wire due to
    an increase in efficiency but don't expect it to be enough to be worth
    changing to a different gauge. I'm just trying to save a buck on
    wire cost.

    I have Tefzel insulated wire I already own . . . . Copper prices
    have doubled in the last year or two. One pound was costing me $12,
    one half pound now costs $35 from the same source - but in larger
    quantities it has (only) doubled. I need about 12 ounces.
  15. I have no experience wiring motors but, it sounds like you think your wire
    will work and you don't want to hear that it won't... It probably will
    work.. All you can do is try it and see.. That is the only way to get a
    definative answer.. I had a similar problem, well similar with everone
    telling me that it wouldnt work. I was going to try to run ethernet over
    700ft of cat5 wire. Everyone said it wouldn't work, but I tried anyway and
    it worked just fine, with the exception of lightning..

    Anyway my point is just do it and see if it works, it probably will...

    - Mike
  16. Jim Land

    Jim Land Guest

    LOL! We should all be so lucky in our projects.
  17. Jamie

    Jamie Guest

  18. Jamie

    Jamie Guest

    when we send out motors for repair we also request them to use teflon
    instead of the various ranges of enamel's they have out there.
    in one location we use AC motors as tension units, they don't last
    long when running at the max with out proper ventilation and cool air
    going across the unit. so we started to have these units repaired with
    teflon and they have been great ever since with the exceptional bearing
    blow out now and then which we can fix our self's.
  19. Jamie

    Jamie Guest

    Hmm, 700Ft?,. bet you r only getting 10MB ?

    and btw, you can get FEP motor/mag wire with
    thin applied coatings on it.
    tefzel, well we had experience with that, it most likely
    will work how ever, if the motor burns up , toxic gas is
    we use that at work for products that need to be
    smokeless. well it mite be Smokeless but it isn't toxless.
    ... if this stuff gets a little over heated you don't want to
    be directly in the path of its venting area.
    FEP isn't much better either actually, that is smokeless but
    removes the O2 of of the air and if your a smoker! watch out.
  20. jasen

    jasen Guest

    All that the inductance can do is smooth out the lumps in what's already
    pretty smooth.
    unless there's extra turns on the output windings we're talking 1 volt
    maximum above what the main output is producing.

    say the thing is going all out and the main rectifier doides are dropping
    1.4v (each) the extra diodes for the field aur under less extreme load
    and maybe dropping 0.6V so you're only 0.8V above the voltage on the output
    RTV silicone may be useful to reduce movement in the wires too,
    that blue stuff used for sealing tappet covers sounds suitable
    (Permatex ultra blue - nuetral cure, heat, and oil resistant)
    Hmm, I wonder how they do that...

    maybe their mail processing hardware bends the letters somewhere along the
    path... or maybe there's someone who sticvks them in a microwave oven :^)

Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day